Assessing the Association Between Blood Loss and Postoperative Hemoglobin After Cesarean Delivery: A Prospective Study of 4 Blood Loss Measurement Modalities.

Background: Visual estimation and gravimetric methods are commonly used to quantify the volume of blood loss during cesarean delivery (CD). However, the correlation between blood loss and post-CD hemoglobin (Hb) is poorly studied, and it is unclear whether the correlation varies according to how blood loss is measured.
Methods: After obtaining Institutional Review Board approval, we performed a prospective study of 61 women undergoing CD to assess the relations between post-CD Hb and blood loss measured using 4 modalities: gravimetric blood loss measurement (gBL), visual blood loss estimation by a blinded obstetrician (oBL) and anesthesiologist (aBL), and the Triton System (tBL). Hb was measured preoperatively and within 10 minutes after CD. gBL was quantified as blood volume in a suction canister in addition to the weight of blood-soaked sponges. tBL was measured with the Triton System by photographing blood-soaked sponges and suction canister contents. To assess the relation between blood loss and post-CD Hb, we performed correlation analyses and compared the magnitude of the correlations across the 4 measurement modalities using William t test. A Bonferroni correction was set to identify a statistically significant correlation (P < .0125) and statistically significant differences between correlation coefficients (P < .008).
Results: The mean (standard deviation) preoperative Hb was 12 (1.1) g/dL and post-CD Hb was 11.3 (1.0) g/dL. Median (interquartile range) values for gBL, oBL, aBL, and tBL were 672 mL (266-970), 700 mL (600-800), 750 mL (600-1000), and 496 mL (374-729), respectively. A statistically significant but weak correlation was observed between tBL and post-CD Hb (r = -0.33; P = .01). No statistically significant correlations were observed among aBL (r = -0.25; P = .06), oBL (r = -0.2; P = .13), and gBL (r = -0.3; P = .03) with post-CD Hb. We did not detect any significant differences between any 2 correlation coefficients across the 4 modalities.
Conclusions: Given that we observed only weak correlations between each modality with post-CD Hb and no significant differences in the magnitude of the correlations across the 4 modalities, there may be limited clinical utility in estimating post-CD Hb from blood loss values measured with any of the 4 modalities.